gtfs_df 0.10.0 → 0.10.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 0be3962480f99e4a2194d6e7da248d5ef82dc5479491ac807211b3dc4d69415a
-  data.tar.gz: ceeaf189058006c0db3b26e6438b8c7fc7a0a077a48f192b588b53e74de695e8
+  metadata.gz: f128bb120303c7d0a4508f0d036812375893ea8c210a351535199b26f3aa3263
+  data.tar.gz: 554926a20907e8dedf02f786daa422a6bce1e3f876074a6b8106682b8a671eeb
 SHA512:
-  metadata.gz: 1f27c1b9493aaf2220dfe9369a06a9c5cc26bcae2b9777b0d33bb061a47e1c176eea309580f893fad53dd80b3e7148e84a65ed6f3e0752ac6203c961c991d4ec
-  data.tar.gz: ad94ff3fa5b2fb69f119fdcac7b8c14df0c2ad7863e42b692d10f5eb0bda151f204618ae8fd7209e9de5ebe0684187530e0e4f789aefaaabf340c8890e9d5df3
+  metadata.gz: 140df9b1b30d5210e6212ee8c7ee158217d26e2605ce9da9d0502eec90777ec4a4ee06a81555d858d69b53614af41d153836532aadd1a0053ec539b8caa9337e
+  data.tar.gz: 9f4fa2528b9897cac3aefb2019b80068adb2fc7022fd66601a91de284b7888cefafc4f5b2ad3f773ab01bb69849f43231fb6ef08155d0bc40fe1bcae9be5a8b1

data/CHANGELOG.md

@@ -1,3 +1,23 @@
+## [0.10.2] - 2026-03-20
+
+### 🐛 Bug Fixes
+
+- Handle nested feeds
+
+### ⚙️ Miscellaneous Tasks
+
+- Raise clear message for multi-feed inputs
+## [0.10.1] - 2026-03-18
+
+### 🐛 Bug Fixes
+
+- Sanitize feed input
+
+### ⚙️ Miscellaneous Tasks
+
+- Normalize platforms
+- Ignore the examples folder on publishing
+- Bump version to 0.10.1
 ## [0.10.0] - 2026-03-06
 
 ### 🚀 Features
@@ -9,6 +29,10 @@
 ### 🐛 Bug Fixes
 
 - Service dates and busiest week fixes
+
+### ⚙️ Miscellaneous Tasks
+
+- Bump version to 0.10.0
 ## [0.9.3] - 2026-02-27
 
 ### 🐛 Bug Fixes

data/lib/gtfs_df/reader.rb

@@ -10,11 +10,30 @@ module GtfsDf
     def self.load_from_zip(zip_path, parse_times: false)
       data = nil
 
+      relevant_files = GtfsDf::Feed::GTFS_FILES
+        .map { |name| "#{name}.txt" }
+        .to_set
+
+      seen = {}
+
       Dir.mktmpdir do |tmpdir|
         Zip::File.open(zip_path) do |zip_file|
           zip_file.each do |entry|
-            next unless entry.file?
-            entry.extract(destination_directory: tmpdir)
+            # Extract files in nested directories into the root of the tmpdir
+            file_name = File.basename(entry.name)
+
+            if seen[file_name]
+              raise GtfsDf::Error, "Found multiple instances of the same file: #{seen[file_name]} and #{entry.name}"
+            end
+
+            # We're skipping:
+            # - unrelated files
+            # - empty feed files
+            next unless relevant_files.include?(file_name) && has_header?(entry)
+
+            seen[file_name] = entry.name
+
+            entry.extract(file_name, destination_directory: tmpdir)
           end
         end
 
@@ -45,5 +64,11 @@ module GtfsDf
       schema_class_name = gtfs_file.split("_").map(&:capitalize).join
       GtfsDf::Schema.const_get(schema_class_name).new(path).df
     end
+
+    private_class_method def self.has_header?(zip_entry)
+      zip_entry.get_input_stream.readline.strip != ""
+    rescue
+      false
+    end
   end
 end

data/lib/gtfs_df/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module GtfsDf
-  VERSION = "0.10.0"
+  VERSION = "0.10.2"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: gtfs_df
 version: !ruby/object:Gem::Version
-  version: 0.10.0
+  version: 0.10.2
 platform: ruby
 authors:
 - David Mejorado
@@ -90,11 +90,6 @@ files:
 - devenv.lock
 - devenv.nix
 - devenv.yaml
-- examples/split-by-agency/.gitignore
-- examples/split-by-agency/Gemfile
-- examples/split-by-agency/Gemfile.lock
-- examples/split-by-agency/README.md
-- examples/split-by-agency/split_by_agency.rb
 - lib/gtfs_df.rb
 - lib/gtfs_df/base_gtfs_table.rb
 - lib/gtfs_df/feed.rb

data/examples/split-by-agency/.gitignore

@@ -1 +0,0 @@
-output/

data/examples/split-by-agency/Gemfile

@@ -1,5 +0,0 @@
-source "https://gem.coop"
-
-gem "gtfs_df", path: "../.."
-gem "optparse"
-gem "whirly"

data/examples/split-by-agency/Gemfile.lock

@@ -1,54 +0,0 @@
-PATH
-  remote: ../..
-  specs:
-    gtfs_df (0.10.0)
-      networkx (~> 0.4)
-      polars-df (~> 0.22, < 0.24)
-      rubyzip (>= 3.0, < 4.0)
-
-GEM
-  remote: https://gem.coop/
-  specs:
-    bigdecimal (4.0.1)
-    json (2.18.0)
-    matrix (0.4.3)
-    networkx (0.4.0)
-      matrix (~> 0.4)
-      rb_heap (~> 1.0)
-    optparse (0.8.1)
-    polars-df (0.23.0-aarch64-linux)
-      bigdecimal
-    polars-df (0.23.0-aarch64-linux-musl)
-      bigdecimal
-    polars-df (0.23.0-arm64-darwin)
-      bigdecimal
-    polars-df (0.23.0-x86_64-darwin)
-      bigdecimal
-    polars-df (0.23.0-x86_64-linux)
-      bigdecimal
-    polars-df (0.23.0-x86_64-linux-musl)
-      bigdecimal
-    rb_heap (1.1.0)
-    rubyzip (3.2.2)
-    unicode-display_width (3.2.0)
-      unicode-emoji (~> 4.1)
-    unicode-emoji (4.2.0)
-    whirly (0.4.0)
-      json
-      unicode-display_width (>= 1.1)
-
-PLATFORMS
-  aarch64-linux
-  aarch64-linux-musl
-  arm64-darwin
-  x86_64-darwin
-  x86_64-linux
-  x86_64-linux-musl
-
-DEPENDENCIES
-  gtfs_df!
-  optparse
-  whirly
-
-BUNDLED WITH
-   2.6.9

data/examples/split-by-agency/README.md

@@ -1,26 +0,0 @@
-# Split GTFS by Agency Example
-
-This example demonstrates how to split a GTFS zip file into multiple files, one for each specified `agency_id`, using the `gtfs_df` Ruby gem.
-
-## Usage
-
-```
-bundle install
-ruby split_by_agency.rb -i <input-gtfs.zip> --ids agency1,agency2
-```
-
-- The output files will be written to the `output/` directory, named `<agency_id>.zip`.
-
-## Options
-- `-i`, `--input PATH` — Path to the input GTFS zip file
-- `--ids IDS` — Comma-separated list of agency IDs to extract
-
-## Example
-
-```
-ruby split_by_agency.rb -i ../../spec/fixtures/sample_gtfs.zip --ids DTA,OTA
-```
-
----
-
-This is a port of the [original Python script](https://gist.github.com/davidmh/f51e5d93a9213e0e606a43167ff87403) using Partridge.

data/examples/split-by-agency/split_by_agency.rb

@@ -1,63 +0,0 @@
-#!/usr/bin/env ruby
-# frozen_string_literal: true
-
-require "optparse"
-require "fileutils"
-require "gtfs_df"
-require "whirly"
-
-options = {}
-OptionParser.new do |opts|
-  opts.banner = "Usage: split_by_agency.rb -i <input-gtfs.zip> --ids NUMBERS"
-
-  opts.on("-i", "--input PATH", "Path to the input GTFS file") do |v|
-    options[:input] = v
-  end
-  opts.on("--ids IDS", "Comma-separated list of agency_ids") do |v|
-    options[:ids] = v
-  end
-end.parse!
-
-unless options[:input] && options[:ids]
-  warn "Both --input and --ids are required."
-  exit 1
-end
-
-input_path = File.expand_path(options[:input])
-agency_ids = options[:ids].split(",")
-output_dir = File.expand_path("./output", __dir__)
-FileUtils.mkdir_p(output_dir)
-
-feed = nil
-
-Whirly.configure spinner: "dots", stop: "✓"
-
-Whirly.start do
-  Whirly.status = "Loading"
-
-  start_time = Time.now
-  feed = GtfsDf::Reader.load_from_zip(input_path)
-  elapsed = Time.now - start_time
-
-  Whirly.status = "Loaded (#{elapsed.round(2)}s)"
-end
-
-agency_ids.each do |agency_id|
-  Whirly.start do
-    output_path = File.join(output_dir, "#{agency_id}.zip")
-
-    start_time = Time.now
-
-    Whirly.status = "-> #{agency_id} filtering..."
-    filtered_feed = feed.filter({"agency" => {"agency_id" => agency_id}})
-
-    Whirly.status = "-> #{agency_id} writing..."
-    GtfsDf::Writer.write_to_zip(filtered_feed, output_path)
-
-    elapsed = Time.now - start_time
-
-    Whirly.status = "-> #{agency_id}.zip (#{elapsed.round(2)}s)"
-  end
-end
-
-puts "✓  Done, all files are stored in the output/ directory"